Correlation Does Not Imply Causation

Here’s a somewhat funny cartoon I came across:

This “correlation does not imply causation” phrase is pretty useful, don’t you think?

“Correlation does not imply causation is a phrase used in science and statistics to emphasize that correlation between two variables does not automatically imply that one causes the other (though it needs to be stressed that it doesn't remove the fact that correlation can still be a hint, whether powerful or otherwise).”

by Wikipedia (linked to page)

So, let me offer my own (messy!) example:

Recently, my sleeping pattern has deteriorated so much that I get to sleep only by 5 or 6 am. 8 out of 10 of my recent posts have also been done in the morning period between midnight and 6 am.

Therefore, A. staying awake after midnight causes me to B. blog. Or so you think!

The quite strong correlation between A and B (me blogging at the period after midnight that I’d stay awake) doesn’t imply that A causes B.

The reason I stay awake a lot is because I do most of the things (besides sleep) I do at home at night. I get so carried away with doing and keeping up with things that I would, over time, get more and more used to sleeping very late.This is also the reason why I blog mostly after midnight as I would usually prefer doing other things first than blogging.

So there you have it! Both A and B are caused by a third factor, C. I have a LOT of things to do/to keep up with, which therefore gives rise to a correlation between A and B.

Does that make sense? Well phew, that was hard! *please don’t think that I was only trying to sound smart* And actually, I should have rephrased that particular bit to, “C. I have a LOT of things to waste time on.”

By the way, you can find more cartoons by the same drawer at XKCD. I haven’t browsed around much, but below are some of the funny ones. If the pictures are too hard to read, simply click on them.